This research has as its objective the development of a practical and precise method for determining absorbed-dose distributions from high-energy electron beams for use in cancer treatment planning, particularly in cases complicated by inhomogeneities such as bone and lung tissue, and by body curvature. This will be accomplished by generalizing Fermi age diffusion theory, already successfully employed in nuclear reactor physics, to the dosimetry of high-energy electron beams. In order to test and further refine the Fermi age model, experiments simulating various inhomogeneities are being carried out. Our final goal is to develop an algorithm for computer calculation of dose distribution from high-energy electron beams, applicable to realistic clinical situations.